Keynotes
Keynote Speakers
Prof Giuseppe Di Fatta
Free University of Bozen-Bolzano, Italy
Title: Multi-Task Deep Learning
This talk provides an overview of Multi-Task Learning (MTL) and to its adoption in Deep Neural Networks (DNN). MTL is based on the introduction of an inductive bias in a learning process when multiple tasks are learned simultaneously. It aims at achieving better performance in jointly learned models than learning each task independently. The underlying assumption is that leveraging on common structure, on shared domain information and, in general, on similarities of related tasks, can lead to better a generalisation level of data models, avoiding overfitting and eventually compensating the scarcity and sparsity of data on each individual task. Moreover, it has been shown that the benefits of MTL can be extended to cases with groups of tasks of unknown relation. MTL can be considered an approach to ‘transfer’ learning, where each task is learned simultaneously, whereas in other transfer learning approaches the models are learned sequentially. MTL for DNN is an interesting research direction, which may help to investigate and understand deep learning processes, and to develop more efficient and more effective ones.
Bio:
Giuseppe Di Fatta (https://www.unibz.it/it/faculties/engineering/academic-staff/person/46582-giuseppe-di-fatta) is a Professor at the Free University of Bozen-Bolzano (Italy) since 2022. He is the Director of the MSc Computing for Data Science and of the executive MSc Artificial Intelligence Applications and Innovations. Previously, from 2006 to 2021 he was with the University of Reading (UK), where he served as Head of the Department of Computer Science from 2016 to 2021. From 2004 to 2006, he was with the University of Konstanz, Germany, where he joined the initial development team of KNIME, a popular data science and machine learning platform. From 2000 to 2004, he was with the High-Performance Computing and Networking Institute of the National Research Council, Italy. In 1999, he was a research fellow at the International Computer Science Institute (ICSI), Berkeley, CA, USA. His research interests and expertise include artificial intelligence, machine learning, data science and data-driven multidisciplinary applications for scientific and industrial domains. He has published over 130 articles in peer-reviewed conferences and journals. He has participated in several research projects, including some funded by the EU Horizon programme. He has been a member of the Institute of Electrical and Electronics Engineers (IEEE) since 2002 and a Fellow of the Higher Education Academy since 2009. He is a member of the Technical Committee on Machine Learning (TC-ML) of the IEEE SMC Society.
Prof Simon Parsons
University of Lincoln, UK
Bio: Simon received his PhD from University of London in 1993, and held academic positions at Queen Mary and Westfield College, Massachusetts Institute of Technology, City University of New York, University of Liverpool and King’s College London before joining Lincoln. At King’s, Simon led the expansion of activity that resulted in the recent re-establishment of a separate Department of Engineering. Simon’s research interests centre on the design and analysis of autonomous systems, in particular with respect to how teams of autonomous systems can decide what to do in complex environments, and how they can explain these decisions.
Prof Aad van Moorsel
Title: Models for Blockchain and Decentralised Systems
We consider blockchains and other decentralised systems from a performance and dependability engineering perspective. The presentation builds on several keynotes delivered in recent years by the author, which consider blockchain performance engineering in the system layer, the consensus layer and the incentives layer, respectively. In this keynote we provide an overall view on model-based performance and dependability analysis and on the software tools that support such modelling. We also discuss in some more detail the BlockSim simulation tool, recent model-based analysis results for the Verifier’s Dilemma and concerns for other emerging decentralised and distributed systems, including those for federated learning.
Bio:
Aad van Moorsel is a Chair in Decentralised Systems and Head of the School of Computer Science at University of Birmingham as well as Director of the institute of Interdisciplinary Data Science and AI in Birmingham and Turing Fellow of the Alan Turing Institute. His research group at University of Birmingham conducts research in security, privacy and trust, with applications in payment, trustworthy AI, blockchain and smart systems. The group’s research contains elements of quantification, be it through system measurement, predictive modelling or on-line adaptation. Aad worked in industry from 1996 until 2003, first as a researcher at Bell Labs/Lucent Technologies in Murray Hill and then as a research manager at Hewlett-Packard Labs in Palo Alto, both in the United States. He got his PhD in computer science from Universiteit Twente in The Netherlands and has a Masters in mathematics from Universiteit Leiden, also in The Netherlands. After finishing his PhD he was a postdoc at the University of Illinois at Urbana-Champaign, Illinois, USA, for two years. He is the author of over 200 peer-reviewed research papers and holds three US patents.
University of Lincoln, School of Computer Science
Bio: Simon received his PhD from University of London in 1993, and held academic positions at Queen Mary and Westfield College, Massachusetts Institute of Technology, City University of New York, University of Liverpool and King’s College London before joining Lincoln. At King’s, Simon led the expansion of activity that resulted in the recent re-establishment of a separate Department of Engineering. Simon’s research interests centre on the design and analysis of autonomous systems, in particular with respect to how teams of autonomous systems can decide what to do in complex environments, and how they can explain these decisions.
